I want to manage sshd's AllowGroups directive and sudo settings with the same defined resource, something like this:

node node1.tld { access { 'group1': sudo => true } access { 'group2': sudo => false } }

So on

node1

I want to allow group1 and group2 to be added to

sshd_config

(or

/etc/sshd_config.d/allowgroups.conf

like this:

AllowGroups group1 group2

I also want to generate sudoers entries.

I think I can deal with sudoers.

I just don't know, how to add properly groups to AllowGroups.

In addition I want to make sure, that if

access { 'group1': sudo => true }

is removed from the node definition, that also renders sshd_config and get

group1

removed.

Any advice, hint or example, how should I do that?

use saz/ssh and saz/sudo

See the

saz-ssh

module at https://forge.puppet.com/saz/ssh?src=slack&channel=puppet

@kenyon as far as I can see, there no option in the saz module to define two groups independently. I can define the list of groups in a single command only.

ssh::server::config::setting

can be defined only a single time for AllowGroups

This message was deleted.

message has been deleted

This message was deleted.

Hi everyone! Here is part of my erb template of iptables (own written module)

<% classes.each do |klasss| -%>
#comment "<%= klasss %>"
<% end -%>

On output i get short list of modules/classes that are on server: #comment "settings" #comment "server-1" #comment "common_servers" #comment "puppet_agent_7" #comment "iptables" #comment "users::user-1" #comment "users::user-2" If on node i look at /opt/puppetlabs/puppet/cache/state/classes.txt I see full ist of classes/modules included in node. How can i get full list via erb template, cause command "classes" seems to be give not full list. At least it does not contain modules that in common_servers, which contain several included modules?

@rismoney do you have your Nomad jobs in a public repository? How do you deal with the jobs? Do you use CI (maybe with Levant?), or puppet, or you run the jobs manually? At the moment I run the jobs manually, but I'm trying to define my best practices.

Hi all , I am trying to install PE 2021 on my local test environment and getting below error on running installer . can someone suggest what i miss here ?

This message was deleted.

I should have used Ceph for the storage, but I'm using Gluster, which is simpler to use. Yesterday I have seen this project regarding postgres: https://github.com/ccakes/nomad-pgsql-patroni

This message was deleted.

For the apache module, what would be the best way to ensure that a module (status and autoindex) are disabled? Looking at the doc there's commands to enable modules but I'm not seeing the opposite? I see there's a a2mod function but it's usage isn't well defined in the readme..? I guess I could run a shell command to a2dismod if the module is present in the enabled folder?

For the apache module, what would be the best way to ensure that a module (status and autoindex) are disabled? Looking at the doc there's commands to enable modules but I'm not seeing the opposite? I see there's a a2mod function but it's usage isn't well defined in the readme..? I guess I could run a shell command to a2dismod if the module is present in the enabled directory?

so right now i do a bit of this- lookup('websvc:classes', Array[String], 'unique').include... in hiera this is websvc:classes is a key with an array of classes to include. like:

yaml
websvc::classes:
  - websvc::whatever
```

if i wanted to declare a class with parameters, based on hiera lookup could i use a resource like declaration? These scenarios i am doing a lookup of a the class like above, but then performing other 'lookups' of separate variables in hieradata. wondering if i could just use the class automatic vars.

like could i do something like-

This message was deleted.

so right now i do a bit of this- lookup('websvc:classes', Array[String], 'unique').include... in hiera this is websvc:classes is a key with an array of classes to include. like:

yaml
websvc::classes:
  - websvc::whatever
```

if i wanted to declare a class with parameters, based on hiera lookup could i use a resource like declaration? These scenarios i am doing a lookup of a the class like above, but then performing other 'lookups' of separate variables in hieradata. wondering if i could just use the class automatic vars.

This message was deleted.

This message was deleted.